aboutsummaryrefslogtreecommitdiff
path: root/nx-X11/programs/Xserver/hw/nxagent/NXdispatch.c
Commit message (Collapse)AuthorAgeFilesLines
* dix/os: backport various signal handling and smart scheduler changes from X.orgMike Gabriel2017-03-211-3/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Backported from X.org: commit 6178b1c91cfc9e860914acc6f0be2f2d2e07a124 Author: Adam Jackson <ajax@redhat.com> Date: Tue Jun 7 15:52:11 2016 -0400 dix: Use OsSignal() not signal() As the man page for the latter states: The effects of signal() in a multithreaded process are unspecified. We already have an interface to call sigaction() instead, use it. Signed-off-by: Adam Jackson <ajax@redhat.com> Reviewed-by: Keith Packard <keithp@keithp.com> commit e10ba9e4b52269b2ac75c4802dce4ca47d169657 Author: Keith Packard <keithp@keithp.com> Date: Wed Nov 11 22:02:01 2015 -0800 Remove non-smart scheduler. Don't require setitimer. This allows the server to call GetTimeInMillis() after each request is processed to avoid needing setitimer. -dumbSched now turns off the setitimer. Reviewed-by: Adam Jackson <ajax@redhat.com> Signed-off-by: Keith Packard <keithp@keithp.com> commit 1f915e8b524dd02011158aa038935970684c7630 Author: Daniel Drake <drake@endlessm.com> Date: Wed May 20 13:16:12 2015 -0600 Keep SIGALRM restart flag after Popen Commit 94ab7455 added SA_RESTART to the SIGALRM handler. However, the Popen code tears down and recreates the SIGALRM handler via OsSignal(), and this flag is dropped at this time. Clean the code to use just a single codepath for creating this signal handler, always applying SA_RESTART. [ajax: Fixed commit id] Reviewed-by: Adam Jackson <ajax@redhat.com> Signed-off-by: Daniel Drake <drake@endlessm.com> commit 94ab7455abc213fc96760e29ab2e943ec682fb22 Author: Daniel Drake <drake@endlessm.com> Date: Tue May 12 16:39:22 2015 -0600 Allow system call restarts upon signal interruption The X server frequently deals with SIGIO and SIGALRM interruptions. If process execution is inside certain blocking system calls when these signals arrive, e.g. with the kernel blocked on a contended semaphore, the system calls will be interrupted. Some system calls are automatically restartable (the kernel re-executes them with the same parameters once the signal handler returns) but only if the signal handler allows it. Set SA_RESTART on the signal handlers to enable this convenient behaviour. Reviewed-by: Adam Jackson <ajax@redhat.com> Signed-off-by: Daniel Drake <drake@endlessm.com> commit a6c71ce5d2d2fe89e07a2ef5041c915acc3dc686 Author: Tiago Vignatti <tiago.vignatti@nokia.com> Date: Mon Mar 28 19:21:28 2011 +0300 os: fix memory and fd leaks in Popen Signed-off-by: Tiago Vignatti <tiago.vignatti@nokia.com> Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net> Reviewed-by: Nicolas Peninguy <nico@lostgeeks.org> commit c9051b684b524549eab6d5b88ee3e195a6f6fbe8 Author: Alan Coopersmith <alan.coopersmith@sun.com> Date: Wed Nov 5 18:25:57 2008 -0800 Use OsSignal in Popen/Pclose to avoid SysV signal() stupidity commit 0e9ef65fa583bf2393dd0fda82df6f092387b425 Author: Keith Packard <keithp@koto.keithp.com> Date: Wed Nov 7 16:33:10 2007 -0800 Don't frob timers unless SmartSchedule is running commit 2338d5c9914e2a43c3a4f7ee0f4355ad0a1ad9e7 Author: Arjan van de Ven <arjan@linux.intel.com> Date: Sun Oct 28 09:37:52 2007 +0100 reduce wakeups from smart scheduler The smart scheduler itimer currently always fires after each request (which in turn causes the CPU to wake out of idle, burning precious power). Rather than doing this, just stop the timer before going into the select() portion of the WaitFor loop. It's a cheap system call, and it will only get called if there's no more commands batched up from the active fd. This change also allows some of the functions to be simplified; setitimer() will only fail if it's passed invalid data, and we don't do that... so make it void and remove all the conditional code that deals with failure. The change also allows us to remove a few variables that were used for housekeeping between the signal handler and the main loop. Signed-off-by: Keith Packard <keithp@koto.keithp.com> **Note**: The above change also required ABI changes in hw/nxagent/. commit abe0a51f3f790f8c055289465e130177c4b647cc Author: Ben Byer <bbyer@bbyer.apple.com> Date: Fri Sep 21 17:07:36 2007 -0700 So, like, checking return codes of system calls (signal, etc) is good. Also, only restore an old signal handler if one was actually set (prevents the server from dying on OS X). commit 6da39c67905500ab2db00a45cda4a9f756cdde96 Author: Eric Anholt <eric@anholt.net> Date: Wed Sep 12 13:23:13 2007 +0000 Fix build on FreeBSD after Popen changes. commit a5b8053606d6e786cdcf6734f271acc05f9cc588 Author: Adam Jackson <ajax@benzedrine.nwnk.net> Date: Tue Sep 11 11:37:06 2007 -0400 Ignore - not just block - SIGALRM around Popen()/Pclose(). Because our "popen" implementation uses stdio, and because nobody's stdio library is capable of surviving signals, we need to make absolutely sure that we hide the SIGALRM from the smart scheduler. Otherwise, when you open a menu in openoffice, and it recompiles XKB to deal with the accelerators, and you popen xkbcomp because we suck, then the scheduler will tell you you're taking forever doing something stupid, and the wait() code will get confused, and input will hang and your CPU usage slams to 100%. Down, not across. Backported-to-NX-by: Mike Gabriel <mike.gabriel@das-netzwerkteam.de>
* replace (DE)ALLOCATE_LOCAL by malloc/freeUlrich Sibiller2017-03-031-4/+4
| | | | | | | | | | | | | | | | | | | | | | This is basically a backport of the following commits + replacing xalloc/xfree by malloc/free. Fixes ArcticaProject/nx-libs#358. commit 2761c103311a1160bc483fd0367d654733df8598 Author: Daniel Stone <daniel@fooishbar.org> Date: Mon Nov 5 14:03:26 2007 +0000 OS: Remove usage of alloca Replace with heap allocations. commit 5e363500c86042c394595e1a6633581eb8fcd1bb Author: Daniel Stone <daniel@fooishbar.org> Date: Mon Nov 5 14:38:28 2007 +0000 OS: Remove ALLOCATE_LOCAL from os.h Remove ALLOCATE_LOCAL_FALLBACK and DEALLOCATE_LOCAL_FALLBACK from os.h, and remove the include of Xalloca.h as well.
* Naming change: Security*Access -> Dix*AccessMike Gabriel2017-02-201-8/+8
| | | | | | | | | | | | Backported from X.org: commit 6c46645cfc1afda8aeabfe0ed4d9342673b702f1 Author: Eamon Walsh <ewalsh@tycho.nsa.gov> Date: Thu Dec 14 14:45:42 2006 -0500 Naming change: Security*Access -> Dix*Access Backported-to-NX-by: Mike Gabriel <mike.gabriel@das-netzwerkteam.de>
* The smart scheduler is not optional.Mathieu Bérard2017-02-151-12/+0
| | | | | | | | | | | | Backported from X.org: commit 9f9268821b13038556fbc029df54ab0e9b2aa77f Author: Mathieu Bérard <mathieu.berard@crans.org> Date: Mon Aug 11 13:52:38 2008 -0400 The smart scheduler is not optional. Backported-to-NX-by: Mike Gabriel <mike.gabriel@das-netzwerkteam.de>
* VCS info lines: Remove ancient X.org / XFree86 VCS info line from code files.Mike Gabriel2016-07-061-3/+0
| | | | | This has already been started while replacing copyright info in file headers and has now been completed with this commit.
* Per-file copyright notices: Update copyright information in file headers ↵Mike Gabriel2016-07-061-7/+15
| | | | that NoMachine placed there own copyright statement in.
* nx-X11/programs/Xserver: Drop {X,x}realloc() macros, use realloc() instead.Mike Gabriel2016-07-021-1/+1
|
* nx-X11/programs/Xserver: Drop {X,x}alloc() macros, use malloc() instead.Mike Gabriel2016-07-021-3/+3
|
* nx-X11/programs/Xserver: Drop {X,x}free() macros, use free() instead.Mike Gabriel2016-07-021-2/+2
| | | | Fixes ArcticaProject/nx-libs#105
* hw/nxagent: Drop code referring to the following removed Xserver extensions: ↵Mike Gabriel2016-06-251-3/+0
| | | | XAPPGROUP, LBX, XEVIE.
* hw/nxagent/NXdispatch.c: Shrink file, drop duplicate code that can ↵Mike Gabriel2016-06-251-3339/+5
| | | | identically be found in dix/dispatch.c.
* Move each screen's root-window pointer into ScreenRec.Mike Gabriel2016-06-211-10/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | Backported from X.org: commit e7fae9ecc42ab5e73b89117722dbf4117d928f9a Author: Jamey Sharp <jamey@minilop.net> Date: Sat May 22 00:26:28 2010 -0700 Move each screen's root-window pointer into ScreenRec. Many references to the WindowTable array already had the corresponding screen pointer handy, which meant they usually looked like "WindowTable[pScreen->myNum]". Adding a field to ScreenRec instead of keeping this information in a parallel array simplifies those expressions, and eliminates a MAXSCREENS-sized array. Since dix uses this data, a screen private entry isn't appropriate. xf86-video-dummy currently uses WindowTable, so it needs to be updated to reflect this change. Signed-off-by: Jamey Sharp <jamey@minilop.net> Reviewed-by: Tiago Vignatti <tiago.vignatti@nokia.com> Tested-by: Tiago Vignatti <tiago.vignatti@nokia.com> (i686 GNU/Linux) Backport to nx-libs: Mike Gabriel <mike.gabriel@das-netzwerkteam.de>
* nx-X11/programs/Xserver: Include nxcomp{,ext,shad} headers like one would do ↵Mike Gabriel2016-06-131-1/+1
| | | | with system-wide shared libraries.
* Rename region macros to eliminate screen argumentMike Gabriel2016-05-021-6/+6
| | | | | | | | | | | | | | | | | | | | | This is a huge mechanical patch and a few small fixups required to finish the job. They were reviewed separately, but because the server does not build without both pieces, I've merged them together at this time. The mechanical changes were performed by running the included 'fix-region' script over the whole nx-X11/programs/Xserver tree: $ cd nx-X11/programs/Xserver && ( git ls-files | grep -v '^fix-' | xargs ./fix-region; ) And then, the white space errors in the resulting patch were fixed using the provided fix-patch-whitespace script. $ sh ./fix-patch-whitespace Thanks to Jamey Sharp for the mighty fine sed-generating sed script. v1: Keith Packard <keithp@keithp.com> (X.Org xserver commit: 2dc138922b7588515d5f2447e4b9dcdc0bef15e0) v2: Mike Gabriel <mike.gabriel@das-netzwerkteam.de> (apply fix-region script to nx-libs)
* Clear header file namespace separation (<X11/...> vs. <nx-X11/...>).Mike Gabriel2015-12-281-5/+6
| | | | | | | | | | | | | | | | | | | | In the process of building nxagent against more and more system-wide installed X.org libraries, we come to the limit of including structs from this (bundled nx-X11) and that (system-wide X.Org) library. This commit introduces a clear namespace separation of headers provided by nx-X11 and headers provided by X.Org. This approach is only temporary as we want to drop all nx-X11 bundled libraries from nx-libs. However, for a while we need to make this separation clear and also ship some reduced fake X.Org headers that avoid pulling in libX* and libNX_X* symbols at the same time. This patch has been tested on Debian jessie and unstable and requires no overall testing on various distros and distro versions, as we finally will drop all libNX_X* libraries and build against X.org's client libs. For now, this hack eases our development / cleanup process.
* Replace 'pointer' type with 'void *'Keith Packard2015-12-281-24/+24
| | | | | | | | | | This lets us stop using the 'pointer' typedef in Xdefs.h as 'pointer' is used throughout the X server for other things, and having duplicate names generates compiler warnings. Signed-off-by: Keith Packard <keithp@keithp.com> Reviewed-by: Eric Anholt <eric@anholt.net> Rebased against NX: Mike Gabriel <mike.gabriel@das-netzwerkteam.de>
* Backport: xserver: Avoid sending uninitialized padding data over the networkPeter Åstrand2015-07-021-1/+19
| | | | Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
* dix: Allow zero-height PutImage requests (fix for X.Org's CVE-2015-3418).pr/dix-cve-fixesKeith Packard2015-05-301-1/+1
| | | | | | | | | | | | | | The length checking code validates PutImage height and byte width by making sure that byte-width >= INT32_MAX / height. If height is zero, this generates a divide by zero exception. Allow zero height requests explicitly, bypassing the INT32_MAX check. Fix for regression introduced by fix for CVE-2014-8092. v2: backports to nx-libs 3.6.x (Mike Gabriel) v3: port to NXdispatch.c rather than dispatch.c (Mike DePaulo) Signed-off-by: Keith Packard <keithp@keithp.com>
* dix: integer overflow in ProcPutImage() [CVE-2014-8092 1/4]Alan Coopersmith2015-05-301-1/+3
| | | | | | | | | | | | | | | | | | | ProcPutImage() calculates a length field from a width, left pad and depth specified by the client (if the specified format is XYPixmap). The calculations for the total amount of memory the server needs for the pixmap can overflow a 32-bit number, causing out-of-bounds memory writes on 32-bit systems (since the length is stored in a long int variable). v2: backport to nx-libs 3.6.x (Mike DePaulo) v3: port to NXdispatch.c rather than dispatch.c (Mike DePaulo) Reported-by: Ilja Van Sprundel <ivansprundel@ioactive.com> Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com> Reviewed-by: Peter Hutterer <peter.hutterer@who-t.net> Conflicts: dix/dispatch.c
* NXAGENT_UPGRADE code cleanup:Mike Gabriel2015-04-161-296/+245
| | | | | | | | | | | NX agent contains/ed two build trees. An old one (probably pre-3.x.y) and a "newer" one. The "newer" code tree used to become enabled by setting NXUpgradeAgentServer in nx-X11/config/cf/host.def to YES. As building the NXUpgradeAgentServer has been the default for years now, we drop all code that does not get used at build time for NXUpgradeAgentServer == YES (i.e., the code that belongs to the pre-3.x.y phase of NX agent).
* Avoid large pixmaps (110_nxagent_createpixmap-bounds-check.full.patch).Mike Gabriel2015-02-101-0/+17
| | | | | | | | | | | | | | | | | It is allowed to try and allocate a pixmap which is larger than 32767 in either dimension. However, all of the framebuffer code is buggy and does not reliably draw to such big pixmaps, basically because the Region data structure operates with signed shorts for the rectangles in it. Furthermore, several places in the X server computes the size in bytes of the pixmap and tries to store it in an integer. This integer can overflow and cause the allocated size to be much smaller. So, such big pixmaps are rejected here with a BadAlloc Originally contributed by FreeNX Team
* Revert "release 3.5.0.19"Mike Gabriel2013-03-281-17/+0
| | | | This reverts commit e77bf36d9afbc7e56522574b06217d57c11dd095.
* release 3.5.0.19Mike Gabriel2013-03-281-0/+17
|
* Imported nxagent-3.5.0-2.tar.gznxagent/3.5.0-2Reinhard Tartler2011-10-101-1/+1
| | | | | | | | Summary: Imported nxagent-3.5.0-2.tar.gz Keywords: Imported nxagent-3.5.0-2.tar.gz into Git repository
* Imported nxagent-3.4.0-5.tar.gznxagent/3.4.0-5Reinhard Tartler2011-10-101-1/+1
| | | | | | | | Summary: Imported nxagent-3.4.0-5.tar.gz Keywords: Imported nxagent-3.4.0-5.tar.gz into Git repository
* Imported nxagent-3.4.0-3.tar.gznxagent/3.4.0-3Reinhard Tartler2011-10-101-1/+1
| | | | | | | | Summary: Imported nxagent-3.4.0-3.tar.gz Keywords: Imported nxagent-3.4.0-3.tar.gz into Git repository
* Imported nxagent-3.4.0-16.tar.gznxagent/3.4.0-16Reinhard Tartler2011-10-101-1/+1
| | | | | | | | Summary: Imported nxagent-3.4.0-16.tar.gz Keywords: Imported nxagent-3.4.0-16.tar.gz into Git repository
* Imported nxagent-3.4.0-11.tar.gznxagent/3.4.0-11Reinhard Tartler2011-10-101-2/+2
| | | | | | | | Summary: Imported nxagent-3.4.0-11.tar.gz Keywords: Imported nxagent-3.4.0-11.tar.gz into Git repository
* Imported nxagent-3.1.0-2.tar.gznxagent/3.1.0-2Reinhard Tartler2011-10-101-0/+4716
Summary: Imported nxagent-3.1.0-2.tar.gz Keywords: Imported nxagent-3.1.0-2.tar.gz into Git repository
generated by cgit v1.2.3 (git 2.25.1) at 2024-11-03 20:06:47 +0000